*   You can now override the health check parameters for AWS ELBs by specifying annotations on the corresponding service. The new annotations are: `healthy-threshold`, `unhealthy-threshold`, `timeout`, `interval`. The prefix for all annotations is  `service.beta.kubernetes.io/aws-load-balancer-healthcheck-`. ([#56024](https://github.com/kubernetes/kubernetes/pull/56024),[ @dimpavloff](https://github.com/dimpavloff))...

- Unhealthy pods covered by PDBs can be successfully evicted if enough healthy pods are available. ([#94381](https://github.com/kubernetes/kubernetes/pull/94381), [@michaelgugino](https://github.com/michaelgugino)) [SIG Apps]

- Fixed a scheduler panic due to internal cache inconsistency ([#71063](https://github.com/kubernetes/kubernetes/pull/71063), [@Huang-Wei](https://github.com/Huang-Wei))
- Report kube-scheduler unhealthy if leader election is deadlocked. ([#71085](https://github.com/kubernetes/kubernetes/pull/71085), [@bsalamat](https://github.com/bsalamat))

- The PodDisruptionBudget `spec.unhealthyPodEvictionPolicy` field has graduated to GA. This field may be set to `AlwaysAllow` to always allow unhealthy pods covered by the PodDisruptionBudget to be evicted. ([#123428](https://github.com/kubernetes/kubernetes/pull/123428), [@atiratree](https://github.com/atiratree)) [SIG Apps, Auth, Node and Testing]

- The PodDisruptionBudget `spec.unhealthyPodEvictionPolicy` field has graduated to beta and is enabled by default. On servers with the feature enabled, this field may be set to `AlwaysAllow` to always allow unhealthy pods covered by the PodDisruptionBudget to be evicted. ([#115363](https://github.com/kubernetes/kubernetes/pull/115363), [@ravisantoshgudimetla](https://github.com/ravisantoshgudimetla)) [SIG Apps, Auth and Node]

- Added a health check for the device plugin gRPC registration server. When the registration server is down, kubelet is marked as unhealthy. If systemd watchdog is configured, this will result in a kubelet restart. ([#128432](https://github.com/kubernetes/kubernetes/pull/128432), [@zhifei92](https://github.com/zhifei92)) [SIG Node]

- Volumes are no longer detached from healthy nodes after 6 minutes timeout. 6 minute force-detach timeout is used only for unhealthy nodes (`node.status.conditions["Ready"] != true`). ([#110721](https://github.com/kubernetes/kubernetes/pull/110721), [@jsafrane](https://github.com/jsafrane)) [SIG Apps]

- Volumes are no longer detached from healthy nodes after 6 minutes timeout. 6 minute force-detach timeout is used only for unhealthy nodes (`node.status.conditions["Ready"]!= true`). ([#110721](https://github.com/kubernetes/kubernetes/pull/110721), [@jsafrane](https://github.com/jsafrane))

### Bug or Regression

- Ensure daemonset controller to count old unhealthy pods towards max unavailable budget ([#127775](https://github.com/kubernetes/kubernetes/pull/127775), [@ncdc](https://github.com/ncdc)) [SIG Apps]

* kubeadm: Detect kubelet readiness and error out if the kubelet is unhealthy ([#51369](https://github.com/kubernetes/kubernetes/pull/51369), [@luxas](https://github.com/luxas))